A Novel Robust Iterated CKF for GNSS/SINS Integrated Navigation Applications

Abstract

In challenging circumstances, the estimation performance of integrated navigation parameters for tightly coupled GNSS/SINS is impacted by outlier measurements. An effective solution that employs a novel iterative sigma-point structure with a modified robustness optimization approach for enhancing the error compensation effectiveness and robustness of filters utilized in GNSS challenge conditions is proposed in this paper. The proposed method modifies the CKF scheme by incorporating nonlinear regression and numerous iteration processes for ameliorating error compensation. Subsequently, a loss function and penalty mechanism are implemented to enhance the filter's robustness to outlier measurements. Further, to fully absorb the valid information of the innovation and speed up the operation of the proposed method, the outlier measurement detection criteria is established to bypass the penalty mechanism against measurement weights in the absence of outliers in GNSS measurements. Field experiments demonstrate that the proposed method outperforms traditional methods in mitigating navigation errors when multipath errors and non-line-of-sight (NLOS) reception are increased.